Whole Genome Sequencing (WGS) for COVID-19 Outbreak Evaluations – How Much Does It Add to Bootstrap Epidemiology & Contact Tracing?

ABSTRACT

Background Epidemiologic investigations are foundational in outbreak evaluations but are unable to fully capture the innumerable interactions that lead to exposure. Whole Genome Sequencing (WGS) offers clonality information that can suggest potential transmission links but is costly and resource intensive. We compared COVID-19 exposure source as determined by contact tracing investigations with linkages inferred by WGS data for a COVID-19 outbreak among healthcare workers (HCWs) and patients. Methods Contact tracing investigations were conducted for HCWs identified in three COVID-19 hospital clusters and included interviews to assess exposure history and infection prevention breaches and categorized as either Community, HCW-to-HCW, Patient-to-HCW, HCW-to-Patient, or Unknown. WGS evaluations were completed for 45 (34 HCWs,11 patients) COVID-19 positive samples (Quiagen EZ1 RNA extraction, Illumina Nextera FLex library/Illumina NextSeq 500). Exposure source determinations were reevaluated using WGS data. Agreement between the two strategies were described as percentage and compared using Cohen's Kappa. Results Among 45 samples submitted, 37 were successfully sequenced, and19 (51%,17 HCWs and 2 patients) were identified as potentially linked clonal Epsilon (B.1.429) COVID-19 variant strains. WGS identified 13 identical and 6 closely related strains that suggested linkages between 15 HCW-HCW, 2 HCW-Patient, 1 community, and 1 unknown transmission. Contact tracing categorized the 19 cases as 8 HCW-HCW, 1 Patient-HCW, 3 Community, and 7 Unknown. After incorporating WGS data, these were reclassified as 9 HCW-HCW, 5 Community, and 5 Unknown. Combining contact tracing with WGS information resulted in 6 (32%) reclassifications;agreement between the two strategies was 58% (Cohen's kappa=0.19), identifying 1 previously unrecognized 1 HCW-HCW and 2 community cases. While contact tracing had suggested 1 patient-HCW transmission, WGS results did not show matching strains. Conclusions WGS can improve the precision of COVID-19 outbreak investigation of transmission links in almost one third of cases.